Adding switches to devo 6s

mwm wrote: You're being confused by the double meaning of the word "pin". The pin numbers in PB's doc are "pins" in the GPIOC port, which are PC10, PC11, and PC12. If you look at the data sheet diagram for pins, you'll see that the pins in the upper left are labeled 78,79 and 80 inside the chip, but PC10, PC11 and PC12 on the outside. So the three GPIOC pins PC10, PC11 and PC12 show up on the chip on pins 78, 79 and 80.

That's what I'm saying; from what I can tell, those pins on the chip, which correlate to PC10, PC11, PC12, don't actually connect to anything! There are no traces coming from them. Nada. ZIp. If you look at the 8s board photos I've linked earlier, those pins are on the "bottom left" of the MCU (pin1 marker is on the "bottom right"), and you'll see a big swathe of nothingness connecting to it.

mwm wrote: Could you check some traces for me?
1) One of those 4 jumper pins is connected to pin 94. The other side of the jumper should be to VCC.
2) What the pins on the other jumper are connected to.
3) TMS is connected to pin 72.
4) TCK to pin 76.

I'll look into it

Just going through my photos, can confirm some things but can't see others enough; will have to get physical

mwm wrote: 1) One of those 4 jumper pins is connected to pin 94. The other side of the jumper should be to VCC.
2) What the pins on the other jumper are connected to.
3) TMS is connected to pin 72.
4) TCK to pin 76.

1. Looks like it goes via an inline resistor? "103", so 10K?
2. Dunno yet. However, in looking at the photos of the back of the main board, it's looking like there might be a bunch of vias under the MCU, so it may well be possible that some of the seemingly unconnected pins are in fact connected to somethings. But only way to test this is to check continuity (or something else), which would involve getting the whole shebang out again; might be a while before I get around to this. So maybe my above comments about PC10-12 are in fact unfounded?
3 & 4. Yep, confirmed.

mwm wrote: There are two jumpers on mine, which I assumed corresponded to the BOOT0 and "power" jumper PB talked about in his "how to flash the bootloader". Could you check some traces for me?
2) What the pins on the other jumper are connected to.

Mike, I assume that's something to do with "power bypass", any clues as to what I should be looking for connection wise? I think they end up at vias under the MCU somewhere, but randomly probing 100 pins isn't high on my "fun things to do" list

RoGuE_StreaK wrote:

mwm wrote: You're being confused by the double meaning of the word "pin". The pin numbers in PB's doc are "pins" in the GPIOC port, which are PC10, PC11, and PC12. If you look at the data sheet diagram for pins, you'll see that the pins in the upper left are labeled 78,79 and 80 inside the chip, but PC10, PC11 and PC12 on the outside. So the three GPIOC pins PC10, PC11 and PC12 show up on the chip on pins 78, 79 and 80.

That's what I'm saying; from what I can tell, those pins on the chip, which correlate to PC10, PC11, PC12, don't actually connect to anything! There are no traces coming from them. Nada. ZIp. If you look at the 8s board photos I've linked earlier, those pins are on the "bottom left" of the MCU (pin1 marker is on the "bottom right"), and you'll see a big swathe of nothingness connecting to it.

Ok, got it. Maybe it's connected by VIA's underneath the mcu? Because the 8 code definitely reads PC10, PC11 and PC12 (FMOD0, DR and MIX2).

mwm wrote: Could you check some traces for me?
1) One of those 4 jumper pins is connected to pin 94. The other side of the jumper should be to VCC.
2) What the pins on the other jumper are connected to.
3) TMS is connected to pin 72.
4) TCK to pin 76.

I'll look into it

Thanks. Nice to know that TMS and TCK are what they say they are. BOOT0 could well go via a resister, but 10K seems a bit high.

I'm hoping that's the other jump is the "power bypass" (PB also calls it the "power override"). Unfortunately, I have no idea what it's supposed to do. I was hoping knowing which pins it's connected to would help with that. Don't bother poking at pins at random! I was hoping to test a bootloader replacement on my 6S, but until someone is willing to test it on a 7E, there's not much point in it.

bitbucket.org/deviationtx/deviation/src/...er=file-view-default

That's pins used for button matrix.

Power bypass/ power override is an 'power off' circuit. Meaning when you turn off tx mcu sees that does little house keeping save setting to eprom and only then does it drops power and tx fully shutdown.

SadSack wrote: Power bypass/ power override is an 'power off' circuit. Meaning when you turn off tx mcu sees that does little house keeping save setting to eprom and only then does it drops power and tx fully shutdown.

RoGuE_StreaK wrote:

SadSack wrote: Once STlink is connected you only need to use jumper for power only.

SadSack, any idea what this jumper connects to / what it does?

My understanding it's a bypass jumper for power. When you switch off tx that tells mcu I want to power down, it then switchs of tx off![/quote]

SadSack wrote: www.deviationtx.com/articles/17-hardware...hardware-connections
Looking there it puts 3.3v to GPIOA:2: output to 0 to shut off Tx

Eek, PA2 is pin 25 ("top right"), definitely a trace coming off it, then goes through a maze of vias, splitting traces, resistors, diodes, capacitors, transistors... Maybe one of the traces manages to go straight to the jumper, I'll give it a test tonight.

Mike, any chance of getting a test build published with the extra trims? No hurry whatsoever, just for testing/curiosity, to see if the theory does in fact relate to reality. The momentary toggles I ordered arrived as non-momentary (ie. the same as a 3-way), so their replacements will be a while getting here anyway, and no guarantee that they'll be correct. I might solder on headers when I get around to it, I'm getting an antenna adapter 3d printed (thanks grnd flyr for the file) to try fitting a 3dBi movable antenna to see if that helps reduce signal loss, so there's a bit of hardware mod to do at some stage. And a short-range nRF24L

One possibly interesting type of toggle switch I've come across is an (on)-off-on; one side is momentary, the other latching. So theoretically you could, with the proper supporting code, use it for say a Gear switch AND a momentary Trainer switch; hold it back for Trainer, to give the student control, release to quickly regain control, and the Master pilot always has control of the Gear switch. Just the first example that came to my head as a possible use for such a setup.

I've attached a test build. It just turns on the L & R trims, and ups the number of trims from 6 to 10 (something I think should be done by default on the 6 anyway - it's not running out of space!).

I'm not getting an ST-LINK connection on the 6, so that doesn't help. Could you confirm that the two jumpers (one above the JTAG points, one near the power outlet) are indeed boot0 and the power bypass.

IIUC, you should only need boot0 if you want to use the embedded bootloader's USART interface, so it's not needed to reflash a fried bootloader with STLINK.

And if PA2 is used to tell the firmware to shut down, then it also shouldn't be needed if you either 1) turn on the power switch, or 2) aren't running the deviation or devention code.

I think I know, but while I'm asking - how did you get the output from the dump_bootloader build? I've got it built, and was waiting to try that (and some other ST-LINK things) before posting. But since I've got the dfu to post..

EDIT: Wouldn't let me post the zip because it was to big. Uploaded test build here: deviationtx.com/downloads-new/category/2...wm-devo6-extra-trims

RoGuE_StreaK wrote: 1. Looks like it goes via an inline resistor? "103", so 10K?

Looks to me that the bottom pin of the jumper does go to "BOOT0" via said resistor.
It would appear that the top pin of the jumper goes through another inline resistor, "472" (4.7K), and on to "PB8", as well as splitting off to other places; must eventually make its way up to the trims.
I can't get definitive reads using resistance check on a cheapie DMM as some pins seem to consistently flicker to a reading before changing to a reading of no continuity.

Thanks for the build, I'll try to get it tested on the weekend.

I finally got my own Devo6S and I traced those connections. I also soldered on a 3-pin header so I can use them later. My multimodule connections worked great, but I failed to anticipate how little room there is in this shell! Good thing I have more parts...

Anyway...I can confirm that the ports for P23 (HOV-T) and P24 (HOV-P) are part of the trim button matrix. The center pin is of both ports is connected together, and connects to the first pin of P18 (the 10-pin connector between the processor board and the smaller board containing the buttons). This circuit attaches to the processor at pin 95 (PB8). The outside pins of P23 and P24 are connected to other buttons in the matrix, as expected. I'll connect some switches and see what if it is fully compatible with the Devo 8 trim configuration.

I found that the center pins from P23/P24 are also connected to the mystery two-pin header near the game port header, through a 4.7k (("472") resistor. If the header is shorted, that line connects to pin 94 (Boot0) through a 10k ("103") resistor.

I was also able to trace the mystery P19 3-hole connection point. The left pin goes to P18-4 then is connected to the processor pin 80 (PC12). The right pin connects to P18-3 and eventually makes its way to the processor, but it disappears into a via underneath the processor. The middle pin is connected to ground. If i read the discussion about what connects to PC10, PC11 and PC12, perhaps I will go back and see if the right pin goes to one of those.

I can post the photos and when I get a chance I'll put them onto my Googledrive.

Edit: P19 does NOT connect to the PC12 pin, I can't tell where it goes. The middle pin is ground, but the other two make it to the 10-pin inter-board connection at positions 3 and 4. From there I can't see the traces. I like the idea of working the processor pins, because we already know most of the connections.

Yeah sorry I haven't gotten around to anything else on this yet, haven't had the time. Hoping to solder on some headers this weekend and test the trims, also solder in a short-range nRF24L module, and try swapping over antennas for a 3dBi.
Unfortunately our local flying field is now buried under spoilage from council construction, so we're battling a bit there to try to secure an alternate field of some sort.

Oh, also got a micro vibrating motor, does the 6 compile have this built-in like the 8S? Does the 8S actually have it included by default now?

Here are the photos of the Devo 6S internals. They should be detailed enough to let you trace many of the circuits.
drive.google.com/folderview?id=0B69zZpwk...kZFhqck0&usp=sharing

I found it very easy to put in the wiring for my TX module updates as they arrive. The multimodule I plan to use worked (with an A7105 attached) but if you build it following PB's instructions it will be too thick to fit. I think all of the TX modules will need to connect through cables as opposed to directly or in sockets.

Edit: is there any way to post these to the downloads section on DeviationTX.com alongside the Devo10, 8 and 7e photos?

Can you edit the wiki? If so, you can add notes to the Devo 8/12 page with those images, or creates a new one as appropriate. If not, I'll do it.

BTW, do keep us posted on putting the mm into a 6!

No, apparently I cannot edit the WIKI. In any case I'll keep notes and update here as I find new stuff.

I am VERY happy to tell you that the HOV-T and HOV-P trim DO WORK!!!! I attached a set of two each momentary buttons to the connections and ran your DFU, and it works great! The trims show up as L+, L-, R+ and R-. Note that these connections are in a very tight space and standard 2mm pins don't really work, as you need to solder FROM THE TOP. I rigged some spare 5-position 2mm female headers I had, but it is a very tight fit! I have added photos to my googledrive with the details.

I still haven't yet figured out P19, but I've wired to the connector now. I am pretty sure that these are actual switches, not momentary buttons, as the center is ground. In the 8, there are 9 switch pins used on GPIOC, and two are not available on the 6, so maybe???:
6: Gear Switch (used on the 6)
7: Elevator D/R Switch - not on the 6?
8: Rudder D/R Switch - not on the 6?
9: FMode 0 Switch (used on the 6)
10: FMode 2 Switch (used on the 6)
11: Aileron D/R Switch (used on the 6)
12: Mix 0 Switch (used on the 6)
13: Mix 2 Switch (used on the 6)

Those switches are NOT visible in the 6 UI, so even if I can confirm their use, I don't think I can test them. Any chance you have another magic build?

I blew out my spare A7105 and the rest of my modules are still on a slow boat, so the MM will have to wait a while. I'm putting the board together with the wiring and I'll let you know as soon as I get a working board!

I'm not going to have time to edit the wiki for a while. Any chance you can edit it in your copy of the repo? Might be better for you to do it for the switch mod in any case. I'm not sure if you can submit a pull request for wiki pages or not. But that will be ideal. If not - I'll add it to the list of things I need to do. Lovely photos, by the way. Like that work surface. Can you tell me what it is?

Good news on the trim switches working! I'll start thinking about how to configure them in hardware.ini. And we may want to use different names, though I'm not sure what they'd be. But I have another change that needs to go in first, so it may take a while.

Ok, since you haven't traced them out, I guess that P19 is connected to the two input pins that the 8 uses for switches that the 6 doesn't, and added support for those. They are set up the same way the 7e extra switches are. You have to enable them in hardware.ini, using either 3x1 or 2x2. 3x2 is ignored. This gives you either SWA0/1/2, or SWA0/1 and SWB0/1.

On the emulator, those show up in the input monitor, but there's no way to change the values, but you ought to be able to check them there. While you're at it, can you tell me what the button monitor shows? I suspect the display may need work.

Test build with both the extra switches & extra trim buttons at deviationtx.com/downloads-new/category/2...devo6-extra-switches

Ok, another trace request. The trainer port plug has connections labelled GND, DECT_RX, TX and NC. I assumed that the NC one wasn't connected, but if it's tied to a pin that would be effing awesome! As in, three rf modules without the MM awesome! I suspect not, but hey - it's worth checking for those of us playing with external boxes.

This is too cool! Yes, P19 has IS CONNECTED TO THE OTHER D/R SWITCHES!

I set up my Hardware.ini to switches = 2x2, and I can see them on the screens just fine. I'll try to get some photos if I can calm down!!!!

It works. Thank you MWM for the support and guidance. And all the other contributors, too. We have two more switches!

I don't have the right switch hardware right now, so I'll get those this week and do some nice photos. I will post updated photos to my googledrive (see earlier in the thread) as I get them.

Thank you once again.

Deal57 wrote: I am VERY happy to tell you that the HOV-T and HOV-P trim DO WORK!!!! I attached a set of two each momentary buttons to the connections and ran your DFU, and it works great!

Dagnabbit, beat me to it! Didn't get a chance to do anything over the weekend.

Deal57 wrote: This is too cool! Yes, P19 has IS CONNECTED TO THE OTHER D/R SWITCHES!

Double dagnabbit!
Should be able to try changing it to 3x1 and see what happens if you toggle one or the other? This should approximate a 3-way switch, essentially two 2-ways back-to-back with a common ground and no chance of both being on at the same time.

I've had a stray though about possibly using momentary trims as toggles, don't know if it's been explored before, will start a new thread and explicate therein...

mwm wrote: Ok, another trace request. The trainer port plug has connections labelled GND, DECT_RX, TX and NC. I assumed that the NC one wasn't connected, but if it's tied to a pin that would be effing awesome! As in, three rf modules without the MM awesome! I suspect not, but hey - it's worth checking for those of us playing with external boxes.

I'd have to take a closer physical look at it, but my photos seem to indicate "unfortunately not"; on the underside it seems that only GND connects to anything (ground plane, funnily enough), and there seems to only be two traces coming out on the top-side. But next time I have it apart I'll have a closer look and a bit of a prod.

I set it up as two switches because I didn't have any 3-way switched around, but of course it would work that way. I think I'll keep them as two because I really want a throttle hold on the left side. So Rudder DR works perfectly for me. With the new trims and switches, this Devo 6S is gonna rock!

I will disassemble the DSC in the morning, if work doesn't get in the way!

Deal57 wrote: I set it up as two switches because I didn't have any 3-way switched around, but of course it would work that way.

I was just saying that with a quick text edit you should be able to check the functionality with your existing hardware, not for permanent usage but just to confirm that it does work both ways.

Deal57 wrote: I really want a throttle hold on the left side.

Ditto! That was a major thing I missed going from the Devo7E, the Mix switch isn't suitable to a quick kill (unless you set both positions for it), and it's not conducive having to reach around for the Gear switch; it's fine for planes for disarming while carrying, not so great for those "oh sh!#" moments with a heli

mwm wrote: I'm not going to have time to edit the wiki for a while. Any chance you can edit it in your copy of the repo? Might be better for you to do it for the switch mod in any case. I'm not sure if you can submit a pull request for wiki pages or not. But that will be ideal. If not - I'll add it to the list of things I need to do. Lovely photos, by the way. Like that work surface. Can you tell me what it is?

Good news on the trim switches working! I'll start thinking about how to configure them in hardware.ini. And we may want to use different names, though I'm not sure what they'd be. But I have another change that needs to go in first, so it may take a while.

Ok, since you haven't traced them out, I guess that P19 is connected to the two input pins that the 8 uses for switches that the 6 doesn't, and added support for those. They are set up the same way the 7e extra switches are. You have to enable them in hardware.ini, using either 3x1 or 2x2. 3x2 is ignored. This gives you either SWA0/1/2, or SWA0/1 and SWB0/1.

On the emulator, those show up in the input monitor, but there's no way to change the values, but you ought to be able to check them there. While you're at it, can you tell me what the button monitor shows? I suspect the display may need work.

Test build with both the extra switches & extra trim buttons at deviationtx.com/downloads-new/category/2...devo6-extra-switches

Ok, another trace request. The trainer port plug has connections labelled GND, DECT_RX, TX and NC. I assumed that the NC one wasn't connected, but if it's tied to a pin that would be effing awesome! As in, three rf modules without the MM awesome! I suspect not, but hey - it's worth checking for those of us playing with external boxes.

Is there a chance we can use different switch names? If so, I suggest using the Rud DR0/1 and Ail DR0/1 names when we use 2 x 2-way switches, to remain consistent with the 8 and 10. For the 3-way switch SWA0/1/2 makes sense. Nice to have.

I traced the trainer port. DECT_RX is connected to the processor pin PA10 through a 100ohm resistor, just like on the 8, and Game_TX is connected to PA9 also through a 100ohm resistor. The problem is that when you put a plug into the jack, DECT_RX is SHORTED to the ground wire, presumably to allow Devo to see that another TX is plugged in. It's something inside the little black box. I tinkered with this on the 7e (same part number) and this behavior is useful for the Arduino add-on because it lets us know not to send PPM out when there is another jack that might be sending PPM In. However, if you wanted to use this port for bidirectional use, you would have to use the same wire for TX and RX.

Unless we change this box! There are two pins that connect to the "ring" of a stereo (TRS) plug, (two different pins connect to the TIP of the cable I usually see on this port) but these two aren't connected to any wiring. So maybe you've got an RX pin?

Let me know how that might be usable. It seems pretty simple to put the jumper in place. Solder the jumper from the little board to the NC white wire (#2, next to black). Then swap out the #2 and #3 wires on the Devo end of that cable. You would then have an RX wire available, right? I'll draw this up when I have a chance.

Ok, that's it, I'm out for the night. It's been a really good day.

Deal57 wrote: This is too cool! Yes, P19 has IS CONNECTED TO THE OTHER D/R SWITCHES!

Yay! As far as I'm concerned, that makes the question of other alternatives for the trim switches moot. We can give the 6 as many switches as the 8 and 10 have, so just leaving them as trim switches is fine.

The extra switch functionality is shared with the 7E, and those are the names it has used for a while. Changing them would break configurations on the 7E, which isn't going to happen. Sorry.

I need to add code so we can ignore the trim switches before submitting a pull request. Probably not going to happen today.

I'll start a new thread on the DSC port usage.